人臉辨識相關研究的重點之一在於如何改善並提升非均勻光源環境下的辨識率。最主要的問題為光源在人臉上可產生劇烈的變化，增加了系統辨別人臉特徵的難度，大幅降低了辨識率。近年來的相關研究著重於光源校正、特徵抽取、與分類器設計，但卻缺乏深入的效能評比。本論文挑出幾項有效的光源校正法與數種局部特徵的抽取，配合支持向量機(Support Vector Machines)分類器進行辨識效能的評估與比較，結論出何種光源校正法與局部特徵的搭配有助於變化光源下的臉部辨識。本論文的另一重點為將前述結論出的方法應用於即時系統之製作，分析相關的問題並提出有效的解決方案。本研究採用Face Recognition Grand Challenge(FRGC)資料庫做為實驗的樣本，此資料庫中每一位對象的影像均包含了不同時間、光源環境及表情變化等因素，提升了本研究的挑戰性。實驗應用前述結論的方法達到85.19%之辨識率，而本方法應用於即時系統之製作也極具競爭力。

Face recognition across illumination is one of the most challenging problems in image-based face analysis. Most research focus on the methods for illumination normalization, illumination-invariant feature extraction, or classifier design, but few compare the performance of different approaches. This research evaluates and compares the performance of a few competitive approaches for illumination normalization and several methods for local feature extraction, aiming at determining an effective approach for face recognition across illumination. Because the other issue of the central concern of this research is the appropriateness of the determined approach in making a real-time system, the methods with high computational cost are excluded, although some may result in high recognition rates. The approach recommended by this comparison study can attain 85.19% in recognition rate on FRGC 2.0 database. With its relatively low computational cost, the approach is experimentally proven appropriate for making a real-time system.

[1] Jie Zou, Qiang Ji, and George Nagy, “A Comparative Study of Local Matching Approach for Face Recognition”, IEEE Transcation On Image Processing, vol. 16, pp. 2617 – 2628, 2007.

[2] Chengjun Liu and Harry Wechsler, Neural Networks, “Independent Component Analysis of Gabor Features for Face Recognition”, vol. 14, no. 4, pp. 919-928, 2003.

[3] Bernd Heisele, Purdy Ho, Jane Wu, and Tomaso Poggio, “Face recognition: component-based versus global approaches”, Computer Vision and Image Understanding 91 (2003) 6–21.

[4] Xudong Xie, Kin-Man Lam, “An efficient illumination normalization method for face recognition”, Pattern Recognition Letters 27 (2006) 609–617.

[5] Xiaoyang Tan and Bill Triggs, “Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions”, IEEE Transcation On Image Processing, vol. 19, no. 6, June 2010.

[6] Wen-Chung Kao and Ming-Chai Hsu, “Local Contrast Enhancement for Human Face Recognition in Poor Lighting Conditions”, IEEE International Conference on Systems, Man and Cybernetics, pp. 277 – 282, 2008.

[7] Weilong Chen, Meng Joo Er, Shiqian Wu, “Illumination Compensation and Normalization for Robust Face Recognition Using Discrete Cosine Transform in Logarithm Domain”, IEEE Transcation On Systems, Man, and Cybernetics—Part B: Cybernetics, vol. 36, no. 2, April 2006.

[8] R. C. Gonzalez and R. E. Woods, “Digital Image Processing”, in reading, 1992.

[9] Y. Adini, Y. Moses, and S. Ullman, “Face recognition: the problem of compensating for changes in illumination direction”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 19, no. 7, pp. 721–732, Jul. 1997.

[10] Robot Vision. MA: MIT Press, B. K. P. Horn, Cambridge, 1986.

[11] Phillips, P.J.; Flynn, P.J.; Scruggs, T.; Bowyer, K.W.; Jin Chang; Hoffman, K.; Marques, J.; Jaesik Min; Worek, W.; “Overview of the Face Recognition Grand Challenge”, IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 947 – 954, 2005.

[12] Xiaoyang Tan and Bill Triggs, “Enhanced Local Texture Feature Sets for Face Recognition under Difficult Lighting Conditions”, Proceedings of the 3rd international conference on Analysis and modeling of faces and gestures, pp. 168-182, 2007.

[13] Xiaoyang Tan and Bill Triggs, “Fusing Gabor and LBP Feature Sets for Kernel-Based Face Recognition”, Analysis and Modeling of Faces and Gestures, Lecture Notes in Computer Science, vol. 4778/2007, pp. 235 – 249, 2007.

[14] Gang Hua, Amir Akbarzadeh, “A Robust Elastic and Partial Matching Metric for Face Recognition”, IEEE International Conference on Computer Vision, pp. 2082 – 2089, 2009.

[15] Kalpana.C. Jondhale and Dr. L.M.Waghmare, “Performance Analysis of DeT in Logarithm Domain and Two -Point Normalization Method for Illumination and Expression Variation in Face Recognition”, IEEE International Conference on Computer Science and Information Technology, vol. 2, pp. 122 – 125, 2010.

[16] Xiaohua Xie, Wei-Shi Zheng, Jianhuang Lai, Pong C. Yuen, “Face Illumination Normalization on Large and Small Scale Features”, IEEE Conference on Computer Vision and Pattern Recognition, pp. 1 – 8, 2008.

[17] Zhiming Liu, Chengjun Liu, “Fusion of the complementary Discrete Cosine Features in the YIQ color space for face recognition”, Computer Vision and Image Understanding, pp. 249–262, 2008.

[18] Ekenel, H.K., Fischer, M., Tekeli, E., Stiefelhagen, R., Ercil, A., “Local Binary Pattern Domain Local Appearance Face Recognition”, Signal Processing, Communication and Applications, pp. 1 – 4, 2008.

[19] Savvides, M.; Jingu Heo; Abiantun, R.; Chunyan Xie; Kumar, B.V.K.V., “Class Dependent Kernel Discrete Cosine Transform Features for Enhanced Holistic Face Recognition in FRGC-II”, IEEE International Conference on Acoustics, Speech and Signal Processing, vol. 2,2006.

[20] Gao; Ekenel, H.K.; Fischer, M.; Stiefelhagen, R., “Multi-resolution Local Appearance-based Face Verification”, Hua International Conference on Pattern Recognition (ICPR), pp. 1501 – 1504, 2010.

[21] X. Tan and B. Triggs, “Fusing Gabor and LBP Feature Sets for Kernel-Based Face Recognition,” in Analysis and Modelling of Faces and Gestures, vol. 4778, 2007, pp. 235 – 249.

[22] M. Savvides, R. Abiantun, J. Heo, S. Park, C. Xie, B.V.K. Vijayakumar, "Partial & Holistic Face Recognition on FRGC-II data using Support Vector Machine," cvprw, pp.48, 2006 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW'06), 2006.

[23] Hazim Kemal Ekenel, Rainer Stiefelhagen, "Analysis of Local Appearance-Based Face Recognition: Effects of Feature Selection and Feature Normalization," cvprw, pp.34, 2006 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW'06), 2006.

[24] 林宗勳, “Support Vector Machines 簡介”